Adding a GTX 680 and water cooling

Build Log:

Time to finalize the X2 system and also do a little experimenting. I ordered a GTX 680 used through eBay as I wanted to test how well it’d work through the PCI-Express extender connected to a PCI-Express 1.0 x1 slot. It replaced the GT 620 for this testing. I mentioned in the previous article that I intended to buy a pair of GTX 680s to make use of the Koolance GTX 680 water blocks I still have.

gtx680

But again, the idea here was testing throughput: would the 1.0 x1 slot constrict communication with the card? For reference, the GTX 770 — which is practically the same processor as the GTX 680 — is topping out at 68 GFLOPS on the Milkyway@Home jobs when plugged into a PCI-Express 2.0×16 slot. Note as well that the GTX 770s have 4GB of memory, whereas the GTX 680 I purchased has only 2GB, the same as the GT 620 and GTX 660s.

So what’re the results?

gtx680-boinc

gtx770-boinc

The GTX 680 actually performs better. On a similar job, the GTX 680 finished about 25 seconds sooner. How can this be?

The only explanation for this is simply the operating system. My FX-8350 system runs Windows 8.1 Pro, which is going to compete with Berkeley for the graphics subsystem. The X2 is being run on a headless Linux server that does not have a graphical windowing system installed, meaning the Berkeley client has full access to the graphics card for processing.

But this shows that for the GTX 680 — meaning also the GTX 770 — the PCI-Express 1.0 x1 lane is not going to be a constriction. I’m now very curious as to how well a GTX 780, or even a GTX 980 would perform on this.

Hell I’d even be willing to test a Titan X on this. Anyone care to lend me one?

Radiator panels

The graphics cards are going to be water cooled in this setup. Eventually. But part of shopping around was trying to figure out how best to do this. I considered a custom enclosure for holding radiators, a pump and reservoir, but decided that would be overkill. And extremely expensive. So then the idea came to mind of just bolting the radiators to a panel with the pump and reservoir either bolted to the radiator or sitting on a shelf behind it. I opted for the latter.

This wasn’t just about saving money but eliminating needless complication. Going with a panel would be simple. An enclosure? Not so much. I’d either have to modify an existing enclosure or have something custom fabricated. Instead I went with a custom cut panel.

Previously I engaged Protocase on having a custom enclosure built. This time, I decided to go across the pond and engage a company in the United Kingdom called All Metal Parts. They have a 3U fan panel for three 120mm fans, so I contacted them about a custom panel for a radiator. Their FAQ page quoted £40 for one panel, £60 for two panels plus shipping.

Specifically I’ll be using the EX360 radiator from XS-PC on this, though I’m not sure yet if I’ll have the fans in push or pull. If I go for push, I’ll probably have the panel sandwiched between the fans and the radiator. All Metal Parts held to the £60 for the panels plus £28 for shipping, which is about $140US total. I was hoping for less since what I originally asked for would’ve been a modification on a panel they already sold.

The design they sent for approval, though, after asking my opinion in advance, has the opening in the panel matching the side of the radiator. In the request, I provided a link to the product page for the radiator, which has a diagram on it, making it easy to design the panel. My only note was to make sure the screw holes were 4mm, large enough to allow a #6-32 screw to pass through.

The desire to water cool this setup is merely about cooling. For one the GTX 660 and 680 sit in the lower 60s C while the Berkeley client is running, and I know that a liquid cooling loop will have them sitting significantly lower — probably even in the 30s C.

boinc-temperatures

Silence is actually not a major consideration in this. As the image above shows, the blowers aren’t spinning much, hovering around 40%, which shows that the cards aren’t being taxed all that much. But it also means there’s not a lot of noise to curtail — the setup shown above actually isn’t all that noisy. As such I’m more concerned about the temperatures.

More proof of concept

Build Log:

The Colony West project continues… this time with more proof of concept on the ideas I have in mind. In this case it’s hooking up the graphics cards through USB panel mount connectors rather than having the USB cables pass through an open slot in the back of the chassis. This test serves two purposes: confirm the idea is feasible, and, in so doing, build out a final (or near final) configuration for this system.

Obviously I needed panel mount connectors (I ordered both 3ft and 1ft lengths, two of each), but along with that I ordered left-angle extension cables — everything being USB 3.0 obviously. This required changing the configuration on the chassis, and I pulled out the Radeon card in the process along with the SATA RAID card. The cables were a little bit of a tight fit, so another set might not work out nearly as well, or I’ll have to get a little more creative. We’ll see.

usbcables

usbcables2

The Fedora 21 install on the SSD didn’t like the changes, so I had no choice but to redo the Linux installation — taking it as a chance to upgrade to Fedora Server 22 in the process.

But the key thing is that the system recognized the graphics cards through the USB panel mount extensions with no loss in performance. While I had little reason to believe this wouldn’t work, again I wanted to be certain. There will be panel mount extension cables on the other side as well, and I have little reason to believe those won’t work as expected. Not sure yet if I’ll do a proof of concept on them, though. Maybe later.

Reply to the New York Times Editorial Board regarding gun laws

I will say up front that I have a concealed carry permit lawfully issued under the laws of the State of Missouri. I’m also up for renewal later this year. Contrary to what many may think, permits are not open-ended. In Missouri they are valid for three years from date of certification by the county sheriff — which is different from date of issue — and there is a statutory 45-day turnaround time while waiting for an extensive background check. If the 45-day time frame expires, the county sheriff must issue a temporary permit that can be revoked if necessary when the background check does come back.

Carrying a firearm concealed without a valid permit is a Class D felony in Missouri.

As I wrote in an earlier article, getting a concealed carry permit is no small endeavor. It’s expensive and time consuming — for Missouri the government side of the coin is $225 to start with a $50 renewal every three years.

So when the idea that a person who wishes to carry concealed must show “good cause” is offered by gun control proponents, in this case by the New York Times, it is about the same as needing to show “good cause” to exercise any of your other rights. Imagine needing to demonstrate “good cause” before the Court appoints an attorney to assist in your defense. Imagine needing to demonstrate “good cause” in saying No to a police search. It’s nonsensical at best, and tyrannical.

Yet I know the gun control proponents want it so that a person must show “good cause” to purchase a firearm. The New York Times stops short of actually saying that in this instance, focusing only on the current case at the United States Court of Appeals for the Ninth Circuit, but I have little reason to doubt they’ve advocated such in the past.

And once again I find myself having to say this: stop using the words “gun safety” to describe gun control laws. “Gun safety” refers to the rules and guidelines regarding the proper way to handle a firearm. The regulations on firearms manufacture that require various mechanisms in the design to prevent the firearm from discharging until the trigger is actually pulled, on the other hand, are gun safety laws and regulations. Laws restricting the size of a magazine or the kind of firearms you can own are not about safety, but control.

And as I’ve pointed out several times on this blog (here and here), gun control laws are racist in their implementation, and gun control proponents are racists, because they will disproportionately impact minorities, since they are more likely to be impoverished, by placing greater cost on exercising their rights.

And having to show “good cause” to exercise a right is a violation of the rights of due process (here and here).

So when the New York Times says that municipalities have a “right and duty…to determine what is necessary for their citizens’ safety”, you know your due process rights are about to be forcefully shoved off a cliff into the La Brea tar pits. You see, due process requires that the government show cause for why I should not be permitted to carry a firearm concealed, instead of the citizen showing cause for why they should be permitted to carry.

Follow-up on ASRock BTC Pro and other options

Build Log:

I’ll just say this up front: the ASRock PCI-E extender can be a little finnicky. For some reason, the GT 620 would stop being recognized by the BOINC client after a period of time, despite the card still being detected by the system. I believe the reason for this was the SATA cables.

I wasn’t using the stock cables that came with the kit. Those are flat, and I was using round cables. But I didn’t initially realize they weren’t the exact same length. So while they looked to be running fine for a period of time, I guess eventually there would be enough latency that the data across the two cables would get out of sync with each other and it’d stop working right. So I changed the cables over to two of the same length, but I didn’t leave the GT 620 connected to it.

gtx660

Instead I connected one of the GTX 660s. And to ensure it was getting adequate power, I made sure to plug it up to a second power supply I had laying around (Corsair GS800) — not entirely sure if it’s completely necessary, but I’d rather have it than not. The card sat at just under 60C with a fan speed of 40%. Having a fan blowing on it — Corsair SP120 with a low-noise adapter — lowered the temperature a little, but also allowed for cool air to mix in with the hot air coming out the back.

The reference blower isn’t the greatest for keeping a graphics card cool — which is why the various brands sometimes make their own coolers for these cards. So when I actually build this out into a final configuration, it’ll likely be water cooled. The GT 620 runs cooler than the GTX 660 — but it’s also not able to process nearly as quickly, so putting a universal VGA block on it probably won’t offer any significant benefit.

But the GTX 660 could pull over 45 GFLOPS consistently according to the BOINC client, well over 2½ times the GT 620. So if I take the two GTX 660s plus the GT 620 together combined, it should get about 110 GFLOPS easily. For those wondering how I’m getting that number, I take the “estimated GFLOPS” for a task and divide it by the total processor time when it completes.

But this also shows that PCI-Express 1.0a x1 — which is a 2 Gb/sec lane — is not a bottleneck in this setup. The USB 3.0 cable can support up to 5Gb/sec, which is enough to handle either PCI-Express 1.0 x2 or PCI-Express 2.0 x1. The two SATA III data cables on the ASRock kit could handle 12 Gb/sec in parallel.

The GTX 660 is performing quite admirably compared to the GTX 770s in Beta Orionis, each of which are on PCI-Express 2.0 x16 slots and powered by an FX-8350. I wouldn’t expect it to perform up to the same level as a GTX 770, but staying above 45 GFLOPS compared to the GTX 770’s 64 GFLOPS average compares quite well to specifications I’ve seen online for both processors. The 770 is basically the same core as the GTX 680, though.

I’d actually be interested to see where a GTX 780 would perform, or even a GTX 980. Would the PCI-Express 1.0 lane be a bottlenec for the latest generation?

But if you want to put GPUs to work for BOINC or something similar, you don’t need the newest generation mainboard and processor — the nForce 500 SLI chipset on the MSI K9N4 SLI mainboard was released in 2006.

USB 3.0 PCI-Express extenders

The USB extenders arrived on Saturday. I was really stoked to use one since they are powered by a standard 4-pin Molex connector, and use a standard USB 3.0 male A to male A plug.

usb1

usb2

usb3

USB cables can be longer than SATA cables. The extenders I ordered come with 1m cables. Most SATA cables included with devices (such as the ones included with the ASRock BTC kit) are 6″ or 9″ at most, though you can buy cables that are longer. The maximum length of a SATA cable is also 1m — the ones I was using for the above experiment were about 18″.

There’s another consideration on this: USB 3.0 male A to female A panel cables.

Imagine this. Two or more graphics cards mounted into a 4U chassis — powered by its own power supply — with a USB 3.0 cable connecting it to the main system. This was the consideration I was talking about previously. This would allow you to create one system in either a 1U or 2U chassis, depending on cooling requirements, that acts as the device host while having any configuration of graphics cards in a second chassis connected only by several USB cables.

Now while similar SATA III brackets do exist, the cable length must still be considered — you are still tied to the 1m length limitation which will include the length of the cables used in the brackets. So if you by two brackets and they each have 12″ cables, you can only use a 12″ SATA cable between them.

So in getting the proof of concept with the GTX 660, I connected the GT 620 to it as well to get an idea of whether I could run the two simultaneously — given the age of the mainboard — and what the BOINC client would do with them. Would it run the GPUs only and not bother with any CPU processing?

usb4

Not quite. It took a little bit for the client to realize that it could run CPU tasks on one of the cores, with the second core being used to coordinate with the graphics cards. Both cards were able to process at expected rates. The GT 620 processed at 18 GFLOPS and the GTX 660 was staying north of 45 GFLOPs. No bottlenecks with either card.

Now three GPUs… not sure how BOINC would handle that with a dual core processor, and I’m not going to try to find out either at this point. The GT 620 is slated for the chassis that will house the FX-8370E.

The GTX 660s will go into a 4U chassis with 14 expansion slots so there is room to grow, and, as previously mentioned, I may water cool them to make use of the blocks I still have while also keeping their temperatures lower.

Next step…

So far everything has been working quite well when it comes to the experiments. I’m pleased with the results, pleased that everything is working as expected, and pleased with the performance I’m getting.

The next steps, though, will move this project in the direction of building this into a rack starting with the cabinet. I have the 20U rails, so it’s now a matter of buying the lumber and necessary hardware and building it.

Additional proof of concepts are needed as well. I need to acquire a couple sets of panel mount USB 3.0 cables (mentioned above) and test them as if that was between the graphics cards and the main system and make sure everything will continue to work as expected. I don’t have any reason to think they won’t work, but I’d rather be safe than sorry on that. If all goes well, then I’ll be acquiring chassis, likely starting with the aforementioned 4U chassis for the graphics cards.

Let businesses discriminate

Here’s an idea… given how much “social progress” is supposed to have been made over the last 50+ years, let’s do this: repeal all the anti-discrimination laws, let all of the bigots come out of the woodwork, and shine the light of day on them instead of forcing them to conceal their bigotry under anti-discrimination laws while they make up other excuses to refuse the business of those they don’t want — you know, just as those who dislike atheists will come up with any reason other than religion to turn us away, or will nitpick over any little thing as an excuse to turn us away.

Take away a bigot’s ability to use their bigotry as a reason to turn away those against whom they’re bigoted and they’ll find some other reason to turn them away that is still legal while concealing the fact the person is a bigot.

Doesn’t it make more sense to reveal the bigots for who they really are?

I mean, if they’re only going so far as to deny products and services, how is it fair to force business owners to provide goods and services to customers they don’t want?

If they are using their bigotry to justify violence against others — including the passage of provisions at the government level since government is institutional violence — then obviously that needs to be countered.

And has been countered. Numerous ways.

Let’s enter a hypothetical here. I run a shoppe and I don’t want to cater to — just for the sake of argument — gays. Now my business existing does not mean everyone with the ability to pay is entitled to the products or services I sell. Inherently I have the right to discriminate against whomever I want for whatever reason I want when it comes to the operation of my business — this includes hiring practices — unless I have entered into a contract that forbids such practices (such as a lease for the business space I rent).

To say otherwise is a violation of my right to transact business however I see fit. And this can be seen in the fact that many favor laws (i.e. violence, since government is institutional violence) that would force me to conduct business with customers I don’t want and hire employees I don’t want.

Setting aside the fact that turning away customers and talented employees would likely be bad for my bottom line, that is a choice I should be freely able to make. But at the same time, the consequences of those choices are also mine to accept — contrary to assertions that I’d be able to act on my bigotry “without consequences”. And if I don’t want those consequences, then it is up to me to change my practices.

Concluding the Supreme Court’s 2014 term

One thing that continually amazes me and pisses me off is how seemingly intelligent people can argue that more freedom is somehow a *bad* thing… And I’m not limiting this to just the many articles that have come out denouncing the recent Obergefell ruling.

At least one thing that can be said about this recent Supreme Court term is that they have slapped the government hard whenever Constitutional freedoms have come into play. They

  • declared unconstitutional a program wherein the Federal government seized without compensation half of a raisin production1Horne v. US Department of Agriculture
  • struck a law in Los Angeles, California, that requires hotels to keep records of patrons on file for 90 days to be searched upon request by law enforcement (I wonder what implications that has on 18 USC § 2257, actually)2Los Angeles v. Patel
  • declared part of the Armed Career Criminal Act to be unconstitutionally vague (and I consider it to be a violation of Ring v. Arizona‘s requirement that all aggravating sentencing factors be put before a jury)3Johnson v. United States
  • declared that a State government does not have to honor any proposed license plate design, as the license plate is considered “government speech”4Walker v. Texas Division, Sons of Confederate Veterans
  • limited applicability of the Controlled Substance Analogue Enforcement Act of 1986, striking a blow to the perpetual “war on drugs”5McFadden v. United States
  • citing the First Amendment, struck a city ordinance that limited the display of signs to specific times and places6Reed v. Gilbert, Arizona
  • citing the Civil Rights Act of 1964, declared Abercrombie & Fitch violated the civil rights of a prospective employee due only to her choice of clothing, which was selected due to her religion7EEOC v. Abercrombie & Fitch
  • affirmed the mens rea principle of criminal law with regard to text published online, requiring that the prosecution show that text that appears to be threatening also be intended to convey a threat, that the mere possibility it could be viewed as threatening is not enough for a conviction8Elonis v. United States
  • limited the applicability of local income taxes to just income earned in that locality9Comptroller of the Treasury of Maryland v. Wynne
  • declared a police officer must demonstrate reasonable suspicion for extending a police stop beyond the initial cause for the detention10Rodriguez v. United States
  • declared a post-conviction GPS monitor to be a search under the Fourth Amendment11Grady v. North Carolina, per curiam, the question of whether it is unreasonable is up to the lower Courts to address
  • declared that Amtrak acts as a regulatory agency when it issues “metrics and standards” (i.e. regulations) regarding the performance and scheduling of passenger rail travel12Department of Transportation v. Association of American Railroads
  • declared that a “tangible object” under §1519 of the Sarbanes-Oxley Act of 2002 is “one used to record or preserve information”, meaning it cannot be applied to fish13Yates v. United States
  • declared that a Muslim prisoner can grow a beard of a reasonable length (in the case in question, it was ½”)14Holt v. Hobbs (unanimous) — which can basically be summarized as “They’re not allowing the prisoner to grow a ½” beard because of an interest to regulate contraband? Are you serious?”

* * * * *

I cannot escape discussion of King v. Burwell here.

The Supreme Court ruled that the judiciary should, in short, overlook any drafting errors in a major piece of legislation simply because the regulatory agency responsible for implementing the law is going off some unspecified legislative “intent”.

The facts of the matter are quite clear. The text of the law does not extend Federal subsidies to the Federal exchange — to say it does means you need to re-learn how to read. The IRS even acknowledged such when writing the initial regulations.15Cannon, Michael F. (2015, March 4) “Seven Things You Should Know about the IRS Rule Challenged in King v. Burwell“. But instead of taking it up with Congress to address the obvious oversight — since many have asserted, though it’s been contested, that Congress intended to extend the subsidies to the Federal exchange — the IRS just wrote the rules to extend the subsidies.

Part of the contest of the noted assertion comes this question: “Why have the States set up exchanges at all when the Federal government can just establish its own exchanges in a State and subsidize it from the Treasury?” The only reason the Federal government would’ve been required to at least give States the option of establishing an exchange is State sovereignty, which is still violated by the Federal exchange anyway.

The implications of this ruling are far reaching. While the ruling does not grant the Executive Branch authority to write whatever rules it wants, it does grant the Executive Branch wide latitude to interpret laws in politically expedient ways. Such wide latitude is contrary to the Constitution, as all Federal power originates with Congress — see Article I, Section 8 of the Constitution — who then directs the Executive Branch through legislation on how to carry out that power.

Further, the Constitution does not allow the Federal government to spend money without Congressional approval, of which subsidies can be considered money “drawn from the Treasury”. Such a restriction on Federal spending has now been weakened so long as the Federal agency in question can cite some “intent” of the law or Congress.

While the case and ruling itself are limited to the Affordable Care Act, do not presume that the law’s proponents and other authoritarian politicians will not mine the decision for statements they can use to justify other policy implementations.

This ruling will have other effects. Reason Magazine, through its ReasonTV production, noted that this could have the effect of permitting States to shut down their own State-run exchanges in favor of going the Federal route. This could lead to only further limited choice with regard to health insurance, completely contradictory to the original promises of the law’s proponents — I’m still waiting for my premiums to go down, as they’ve actually doubled over the last 5 years.

* * * * *

In a case in which I’m mixed, Heien v. North Carolina, the Court upheld the validity of a police stop based on a police officer’s mistaken understanding of a law regarding vehicle brake lights. The stop led to a consented search of the vehicle, wherein the officer found cocaine, leading to Heien’s arrest and subsequent conviction.

The concern and controversy with this decision lies in the fact that the law that prompted the stop was somewhat ambiguous. And while the ambiguity of the law was resolved by the Courts — meaning you only need one functioning brake light — there was still an adjacent law requiring that all originally installed rear lights be in working order. While the officer cited the brake light as the reason for the stop, the stop was still reasonable under North Carolina’s statutes, making the subsequent consented search still valid under the Fourth Amendment.

The brake light citation itself could be reasonably challenged and vacated, but because the vehicle was still not operating within what is required by law, the stop is still reasonable. This is line with other standing precedent regarding officers being sued for violating a person’s rights. Typically an officer is allowed general leeway to be reasonably mistaken about how the law is interpreted, but the law does not protect those who are clearly incompetent or knowingly in violation.

I’ve seen interpretations of this decision as allowing officers to be ignorant of the law, and that is not the case. And it certainly brings up one very clear point as well: never consent to a search of your vehicle. And as the Court ruled in another case in this term, if you don’t consent to the search, the officer must demonstrate cause for that search, and the extra time detaining you for it.

* * * * *

Another case that also garnered some controversy is Integrity Staffing Solutions v. Busk. This is the case regarding Amazon’s security procedures for individuals leaving an Amazon warehouse.

The opinion in the case said, to summarize, that the time between clocking out of your shift and getting to your car is not wage time. What made the Amazon case unique is that every person who exits the building is subject to a security search that could take up to an additional half hour of your time. The Supreme Court upheld the conclusion that the employees are not entitled to additional wages for that time, at least under the text of the Fair Labor Standards Act.

This is one of those cases where the Court applies the law, but Congress is still free to act. Many thought the Court should’ve ruled that the time is entitled to pay. Nothing in the law prevents an employer from making that time payable, they’re just not required to do so, so says the Supreme Court. Congress is still free to expand the Fair Labor Standards Act to make that time payable, so lobby your Senators and Representative if you feel the law should be expanded.

This is somewhat contrary to King v. Burwell as well. In Integrity Staffing, the Court applied the law as it was written, but in King, the Court basically had to “go out on a limb”.

One thing that was fun, though, was watching all of the various commentators and bloggers online continually writing articles saying “This case shows how the Court will rule in King” or “This case shows how the Court will rule on gay marriage”. So many facepalms all around in that regard.

References[+]

Adding the GT 620

Build Log:

In the previous article in the series on this project, I mentioned I had a spare GT 620. The card specifically is the Zotac GT 620 2GB. Unfortunately I couldn’t get it into the currently running system, but in making some considerations for additional hardware, I did come up with something else.

One product I’d seen on the shelves at Micro Center is the ASRock BTC Pro. For those unfamiliar with what it is, it’s a PCI-Express extension system — a PCI-E x1 “card” goes into the PCI-E slot on the mainboard, while a PCI-E x16 receiver goes onto the card. It has a custom 5-pin power cable to transfer power between the mainboard and graphics card, and it uses 2 standard SATA cables for data. Pretty nifty little thing.

And I used it to plug the GT 620 into the X2 mainboard to see how well it would work. And it’s working fairly well.

zotac

asrock_btc

Milkyway@Home is reporting the card functioning at about 18 GFLOPS, about 9x the performance of one of the X2’s cores, but — unsurprisingly — about 28% of the performance of one of the GTX 770s in Beta Orionis.

As the pictures above demonstrate, the card is just sitting out in the open currently — this is just a proof of concept more than anything else at the moment. And the temperature reported by the card is just under 50C (fresh thermal compound might do better), so it’s staying well around a decent temperature. The interesting thing is that since part of one CPU core is dedicated to talking with the GT 620, only one core is being used for processing, keeping the temperature inside the case significantly cooler: the cores are sitting under 40C and the chassis is reporting an internal temperature of 37C.

The custom power cable and use of SATA cables doesn’t really impress me, though, so, again, consider this setup just a proof of concept. Searching online I did find similar setups that use a USB 3.0 cable instead — male A to male A. This got some gears turning in my head as to how I could create a setup. USB 3.0 cables can be longer than SATA cables, plus the USB options have a standard 4-pin Molex or 4-pin floppy power connector, meaning they can be powered directly from the power supply.

Again, seeing these got the gears in my head turning, and I’ve got some more experimenting to do as a result. This also provides interesting implications for what to do with regard to future upgrades.

But first, I’ll want to test one of the GTX 660s to see what kind of performance it can provide. I just need to pull the water block off one of the cards to put the stock cooler back on.

Colony West Project

Build Log:

In an earlier post I wrote about my plan to put a bunch of older hardware I have just laying around back to work. As I intend this to be an ongoing, evolving thing, I’ve decided to dub it “Colony West”. Now this isn’t just some name I came up with out of the blue — it’s actually a fictitious name I’ve had registered first with Polk County, Iowa, and now with the Missouri Secretary of State.

Before going into this full-tilt, I needed to have an idea of what I’d be in for.

I already had an Athlon X2 system running. It was the gaming server I built for my wife with the X2 4200+. I installed BOINC to it and have been running it for over 24 hours continuously as I write this so I’d have an idea of what to expect. With both cores maxed out for that entire time, the CPU core temperatures stayed around the mid-40s Celsius. The CPU is being cooled by a Noctua NH-L9A CPU cooler, and the system has three intake fans: two Noctua 60mm fans and an Enermax 80mm fan. The chassis reports an internal temperature of 39C.

So the system is definitely having no difficulty remaining stable and at a good temperature doing this. Currently the CPU time is being contributed to Milkyway@Home, chugging along at about 2.2 GFLOPS per core. My FX-8350, however, is getting close to 6 GFLOPS per core.

So I’m thinking this X2 4200+ system will stay running BOINC, and I’ll take the other X2 and use it for my wife’s Minecraft server. It’s a slightly slower processor, but the mainboard has 4GB of RAM on it. For some reason the MSI board won’t recognize the 4GB of RAM, but the Abit board has no issue with it. Oh well. The Minecraft server will benefit with the extra RAM more than BOINC. Plus that other X2 was my wife’s previous system, so it only seems fitting.

But the rest of the hardware will be purely running BOINC. I’m just not sure which system I’ll bring online next — likely transplant the running system into a 3U pr 4U chassis so I can use an AIO with it along with a better graphics card.

The current build for the X2 4200+ has a Radeon HD5450, which Milkyway@Home does not use — it only uses nVidia CUDA on Linux. I have a spare GT 620 I can put into this. I’ll have an extra cooling consideration, and it’s going to be fueling my decision on which chassis I buy next — I’m thinking a 3U chassis instead of the slim 2U.

I also have a couple GTX 660s still lying around too that I can possibly put into this — but that’ll likely happen when I get the 990FX mainboard up again. I’ve got a couple ideas in mind for that. Speaking of, I’ve opted for the FX-8370E for that system instead of the Opteron.

So given that my experiments have yielded desirable results, time to see where this’ll go. Next step, though, is building the cabinet.

On the Confederate flag

Following the recent mass homicide in South Carolina, attention has turned once again to guns. But it’s also turned to the Confederate flag. Mass calls are coming from within South Carolina and without for the flag to be removed from the state legislature grounds in Charleston.

Here’s an idea: ask the people of South Carolina what they want. Since the legislature is supposed to represent the people, put it up for a popular referendum. Simple question on the ballot: “Do you wish to see the Confederate flag removed from the State legislature grounds?” Yes or no, with abstentions counting toward “no”.

Then once the people of South Carolina have had a say on the matter, regardless of the direction they choose to go, the rest of the United States can just shut the hell up about it. There are more important things that require public attention and scrutiny than a piece of cloth flying over a Confederate memorial in South Carolina.

Putting older hardware to work

Build Log:

If you’re like me, you’ve probably collected a bunch of older hardware that is left over from previous upgrades. And it’s probably sitting around doing nothing. I’ve decided to change that, at least in my instance.

In getting into rack mounting hardware, I figure what I need to do is rack up some of these older systems and put them to work. To that end, here’s the hardware in question:

  • AMD Athlon XP 2500+ with Abit NF7 nForce2 mainboard, 512 MB RAM
  • AMD Athlon 64 2800+ with ASUS K8U-X 754 ULi M1689 mainboard, 512 MB RAM
  • AMD Athlon X2 3800+ with Abit KN9 Ultra mainboard with 4GB RAM

The systems should all fit into 2U server chassis, such as the one used for the server I put together for my wife. The downside is that those are $100 each plus ground shipping, so I’d be looking at just $300 for the chassis plus shipping on all of them. Add in the power supplies and silent fans and things will add up quick. Hopefully I’ve got plenty of spare graphics cards laying around for this as well.

I’ll get started on building these systems once I have a rack cabinet built for all of this, and another rack surge suppressor as well.

So now the bigger question: why? This is all hardware that’s been sitting around — in some cases, for a long, long time. So it is computing power that could be put to use and actually serving a purpose, even if it isn’t my purpose.

I’m talking about distributed computing projects like BOINC, Folding@home, and distributed.net. While the systems mentioned don’t exactly have a lot of computing power to contribute to these causes, it’s better than just sitting around. If the hardware is still viable, I want to put it to use — though given the XP is only a 32-bit processor, it’ll be interesting to see how well it can be put to use.

I also have a 990FX mainboard I’m considering adding into this, along with two spare GTX 660s that I can add in as well. I have a 4U rack chassis that can support that mainboard easily along with both graphics cards, or I’ll go with something shorter. All I need to use that mainboard is just another processor. The question is whether to go with an Opteron or FX. The former would be preferred if I can find one as it’ll have a lower power consumption, but I can go with an FX-8370E if need be.

The plan as well is for the rack to evolve over time. When Absinthe or Beta Orionis are upgraded, those mainboards and processors will replace one of the existing systems in the rack so there is only ever just three or four systems to keep power consumption down — I don’t need to be maxing out a 20A circuit doing this.

But everything first starts with the rack cabinet. I’m probably not going to just stack a set of IKEA RAST nightstands, but will probably build up something — likely similar to the plans over on Tom Builds Stuff. I do need to consider ventilation and cooling on this, especially since we’re talking about systems that will likely be running around the clock. Speaking of, that’s going to mean an interesting consideration for the Athlon 64 (Socket 754) and XP (Socket A) processors and trying to find quiet, lower profile CPU coolers that’ll mount onto their respective sockets.